1. Field of the Invention
The present invention relates to an image processing technique that enables replay after changing the focus state of a moving image.
2. Description of the Related Art
In recent years, imaging apparatuses referred to as a “light-field camera” have been practically implemented. The imaging apparatus divides incident light into multiple light components by a micro lens array arranged on an image sensor so that the imaging apparatus can obtain light beam information by capturing light traveling in a plurality of directions. After shooting, the imaging apparatus executes predetermined calculation processing based on the intensity of light and the light incident direction so that an image with any focal distance and any depth of field (hereinafter collectively referred to as “focus information”) can be configured. The advantage of the light-field camera is that any image can be configured by executing calculation processing after shooting by utilizing light beam information (hereinafter, referred to as an “LF camera”). There are similar advantages during moving shooting using an LF camera, and reconstructing image data during replay after moving shooting and executing moving image replay that reflects arbitrary focus information are possible.
Conventionally, a technique in which a separate image is displayed during fast forward of a moving image is known. In the technique disclosed in Japanese Patent Laid-Open No. 2011-97451, when carrying out fast forwarding of a 3D (3-dimensional) moving image, a 2D (2-dimensional) image is displayed in order to improve viewability. Japanese Patent Laid-Open No. 2009-159357 discloses that after shooting, the depth of field of the image can be adjusted, and the blur state can be adjusted.
However, with respect to moving images captured by the LF camera, in the case in which a user searches for a desired scene by executing fast forward replay and the like, overlooking a scene can occur due to the setting of the focus information. This is because, in the case in which a user is viewing a moving image in a state of focusing on a particular depth in the image, recognizing a subject outside the focus area is difficult. Thus, the user must also pay close attention so as not to overlook a desired scene in an area that is not in focus.
In Japanese Patent Laid-Open No. 2011-97451, although a 2D image is displayed in order to improve the viewability during fast forward playback, recognition of a subject outside a focus area is not taken into consideration in an image for which arbitrary focus information can be specified. In a method of switching from a 3D display to a 2D display, because the focus information itself does not change, the problem of recognizing of a subject outside of the focus area cannot be solved.